Adaptive Login Form: Adjusting compromise between Comfort and Paranoia.
If my current IP address is not marked as Dangerous since my last successful login, then there is no need to distrust me and force me to go through Quests to solve different types of Captchas.
In this case, the standard “Password” field is sufficient for one attempt.
But if the Attempt is unsuccessful, then we mark the IP address as Dangerous, and then it is possible and necessary to trick me (or the one who is trying to be me) with a more thorough login procedure.
There may be multilevel options. It doesn’t matter (this will be gradually added to the functionality). We are now talking about the General Principle.
Separate statistics are generated for each IP address and the ratio “Successful number of entries” / “Total number of entries” is determined. Depending on how close this parameter is to 100%, we can talk about the need for the Toughness of the Mistrust process.
This mechanism starts before the User enters his Login.
The more Unsuccessful Login attempts occur from a given IP Address, the more thoroughly it is checked.
Conversely, the Login procedure can be simplified as much as possible if there is no obvious reason.
- Regardless of what kind of Authentication Error occurred, be it:
- Invalid Username;
- Invalid User Password;
- Incorrectly specified additional security elements: “Secret Key” / Captcha / etc.
This will not be indicated in the error message. There will always be only one message: “Authentication Failed”.
Thus, we do not explicitly indicate to the potential Villain / Bot the reason for the denial of access. And the more such Reasons there are, the more complicated the Entry procedure becomes.
- Integration with “New Users Monitor”